The ability of the DIAGNOdent to detect and quantify enamel demineralisation in various laboratory created lesions was assessed and results indicated that the device was unable to reliably effect this. Comparison was made with another fluorescence based caries detection device which demonstrated a superior performance in this respect. The influence of exogenous staining on the DIAGNOdent analysis of the same laboratory created lesions was investigated and was found to have a profound effect on the instrument’s ability to detect and quantify the degree of demineralisation. Subsequent work, however, determined that the effect of the exogenous staining was unstable beyond an initial period of observation. Further studies examined the ability of the DIAGNOdent to detect residual caries at the margins of amalgam restorations and results using this model indicated that the use of the DIAGNOdent offered no significant advantage over visual examination alone. In a further experiment, the distribution, within natural carious lesions, of the fluorescence measured by the DIAGNOdent was investigated and the DIAGNOdent assessment of the lesion related to a visual assessment of caries extent. Both transverse and longitudinal mapping of fluorescence distribution within the lesion. In addition, no significant correlation was found between the visual and DIAGNOdent assessment of the lesion. The results of these experiments question the ability of the DIAGNOdent to detect caries by recognising demineralisation and would seem to suggest its mode of action to be related to detection of other fluorophores within the lesion. The lack of relationship between the DIAGNOdent and clinical assessment of a lesion along with its apparent limitations in respect of caries detection at restoration margins urge caution in using the device clinically.